Since you insist, I will cave.
If I have a large number of binomially distributed events, I can use the central limit theorem to assess them with a normal distribution and give meaning to the standard deviation.
Thought you had me there, didn’t ya? I’m not entirely fuddy-duddy.